- 23rd May 2019
- Posted by: Manolis
- Category: Big data
From loss prevention to predicting seasonal needs, big data is helping businesses in various industries improve their inventory management practices.
Big data allows for rapid analysis of vast amounts of information at substantially faster speeds than humans could achieve prior to technological developments. Here are six ways it has helped companies excel in the all-important task of inventory management.
Predicting seasonal needs
Retailers and manufacturers know different seasons bring different consumer demands. Before big data platforms became widely available, companies relied on historical data and educated assumptions to decide how to accommodate seasonal fluctuations. Now, big data can reveal specific trends to optimize stock management.
For example, a platform from IBM shows how weather impacts shopping habits. The data indicated that when temperatures get warmer than 77 degrees Fahrenheit, people are less likely to buy ice cream at the grocery store because they fear it will melt. Similarly, when temperatures reach 68 degrees or above in Scotland, barbecue sales triple.
Big data can also help companies plan for seasonal periods or holidays, such as Christmas and back-to-school periods. Brands can look at data from previous years and calculate which specific factors could result in an increased or decreased need for certain items.
Streamlining replenishment needs
Excessive amounts of stock often lead to waste or lost profits due to having to sell things at drastically reduced prices. On the other hand, if customers visit a store and find it consistently does not have the products they want or need, they will get frustrated and may start shopping elsewhere.
Many big data solutions allow companies to decide precisely how much inventory they need to keep customers happy and avoid having too much stuff on hand. Some even have automated features that enable quick reorders.
In one case study involving a discount retail chain with 2,500 stores, employees at the retail outlets spent approximately four hours per day manually checking the shelves and making subjective judgments about what to reorder from the distribution centers. Additionally, the brand did not have a single process for handling replenishment needs across its locations.
However, the chain adopted an automated replenishment solution that allowed using a tool to make weekly sales forecasts at the stores, plus give distributors recommended purchase order lists. Taking this approach resulted in a 38% decrease in out-of-stock events and allowed store employees to spend approximately 30 minutes confirming which items to reorder after looking at the suggestions from the big data platform.
Reducing the reach of recalls
When companies get word of product recalls, they have to act fast to minimize the damage caused to people, stores or companies. One of the easiest ways to do that is to use lot number-tracking software once the brands identify the problematic batches. Lot numbers get assigned to shipments of products, such as all those produced at a particular plant on a specific day.
Then, companies can see which stores received the recalled products, plus which manufacturer produced them. Amazon is an example of a company that uses big data to determine if it needs to stop shipping some products due to possible food safety risks. It monitors 67 public websites daily for incidents of food safety warnings and potential recalls.
Other brands use big data to mine through customer feedback received on social media channels or through a company’s customer service email address. For example, a flurry of comments from people who became ill after eating a particular kind of peanut butter might trigger a data analysis tool to recommend brand representatives investigate the complaints more thoroughly.
Getting insights about parts-related data
Using big data is also advantageous when assessing the various parts that comprise the things people buy. Having an adequate number and type of products on hand is essential if companies offer purchase protection plans that give people the option of extending their item warranties for an additional two years, for example.
Then, the collected data extends beyond the product brands to the service workers handling repairs for customers. In the instance of a clothes dryer with a faulty heating element, the technician might use an app to tell the appliance brand about the issue. The brand could then feed the information into a big data platform and learn that a particular model is more likely than others to have heating element problems.
That issue might not result in a recall. However, it would signal to a brand that it needs to have enough parts in stock to service the existing machines, treating each of their operational areas as separate to ensure that the right components are in the right areas of the country or world. Moving forward, the brand may want to consider looking for another supplier or making a design change to solve the heating element failures.
Enabling cross-selling to encourage more product movement
One of the problems with many online stores is that the inventory they offer is so vast that customers can easily miss opportunities to buy additional products which complement their purchases, even though the e-retailer supplies them. However, it is common for many brands to make suggestions based on a person’s actions during a current visit or their past purchases.
If companies plan carefully, these recommendations can help them move stagnant products. Consider a customer who buys a refurbished iPhone 6S, an older model. Accessories for the 6S might not be advertised in top sellers lists because they are no longer in high demand, however e-commerce executives could use a big data platform to determine which iPhone 6S products it has in inventory and which were historically the biggest sellers, in order to recommend them to the customer buying the refurbished phone.
Enhancing loss prevention strategies
According to a 2018 study, shrinkage costs retailers an average of 1.33% of sales. It happens when inventory reports do not match up with the products on hand because of incidents employee theft or shoplifting incidents with customers.
Fortunately, big data can facilitate improvements in loss prevention. An analytics program might show that a high percentage of shrinkage occurs on Saturday afternoons in the cosmetics department of a chain store, then install more security cameras in that area and increase the security staff members working on Saturdays.
By evaluating transactional data, a store could uncover instances of dishonest employees working together to steal things. A more innocent example of store loss is if someone working in a stockroom misplaces a newly arrived shipment. Big data platforms help brands see which products result in the most substantial losses, promoting changes in processes or resources to cut shrinkage rates.
Ensuring more inventory successes
Some people think of inventory management only in terms of keeping shelves full. However, it also extends to being aware of where products exist in the marketplace and taking the appropriate actions to limit negative consequences, such as shrinkage, product recalls or large volumes of outdated stock. These examples show how big data can help companies make inventory management gains while reducing staff costs and maintaining reputations as reliable, high quality retailers.